rasdani

--headful

Make a web browser multimodal, give it eyes and ears.

chat-your-code

Ask questions about your codebase using GPT

cheatcode

your cheatcode for productivity

distilabel

Distilabel is a framework for synthetic data and AI feedback for AI engineers that require high-quality outputs, full data ownership, and overall efficiency.

dotfiles-v1

germanrag

GermanRAG - a German dataset for finetuning Retrieval Augmented Generation

inference-is-all-you-need

mp-transformer

Learn latent primitives of human movement.

smolR1

reproducing DeepSeek R1 Zero with Qwen2.5-0.5B on two 4090 GPUs

rasdani/smolR1

reproducing DeepSeek R1 Zero with Qwen2.5-0.5B on two 4090 GPUs

rasdani/atropos

Atropos is a Language Model Reinforcement Learning Environments framework for collecting and evaluating LLM trajectories through diverse environments

rasdani/build-smolGRPO

rasdani/dotfiles

managed by chezmoi

rasdani/evalchemy

Automatic evals for LLMs

rasdani/genesys

rasdani/gpt-oss

gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

rasdani/InternVL

[CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的开源多模态对话模型

rasdani/j1-micro

j1-micro (1.7B) & j1-nano (600M) are absurdly tiny but mighty reward models.

rasdani/lerobot

🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

rasdani/modal-examples

Examples of programs built using Modal

rasdani/nano-vllm

Nano vLLM

rasdani/open-instruct

AllenAI's post-training codebase

rasdani/OpenHands

🙌 OpenHands: Code Less, Make More

rasdani/prime

prime is a framework for efficient, globally distributed training of AI models over the internet.

rasdani/prime-cli

The Prime Intellect CLI provides a powerful command-line interface for managing GPU resources across various providers

rasdani/prime-rl

rasdani/R2E-Gym

Official repository for R2E-Gym: Procedural Environment Generation and Hybrid Verifiers for Scaling Open-Weights SWE Agents

rasdani/ray

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

rasdani/reasoning-gym

procedural reasoning datasets

rasdani/rllm

Democratizing Reinforcement Learning for LLMs

rasdani/simpleRL-reason

This is a replicate of DeepSeek-R1-Zero and DeepSeek-R1 training on small models with limited data

rasdani/SkyRL

SkyRL: A Modular Full-stack RL Library for LLMs

rasdani/SWE-smith

Scaling Data for SWE-agents

rasdani/triton

Development repository for the Triton language and compiler

rasdani/unsloth

Finetune Llama 3.3, DeepSeek-R1 & Reasoning LLMs 2x faster with 70% less memory! 🦥

rasdani/verifiers

Verifiers for LLM Reinforcement Learning

rasdani/verl

verl: Volcano Engine Reinforcement Learning for LLMs

rasdani/verl-internvl

rasdani/vllm

A high-throughput and memory-efficient inference and serving engine for LLMs